088 MINERALS AND VEINS OF .EGIR1NE-SYENITE, 



being composed of several kernels, one within the other, of 

 slightly varying composition, as evidenced by progressive differ- 

 ences in physical properties. 



Discordance amongst the several analyses quoted by Dana in 

 respect of the relative proportions of monoxide to sesquioxide 

 bases was sufficient inducement to cause me to make several 

 extra determinations of ferrous iron; these were done on small 

 chips readily obtained from some of the larger crystals showing 

 in the rock. As a result the FeO constituent was found to vary 

 considerably, up to almost 8%, and in one case completely absent; 

 in this last assay the chip had the appearance of being specially 

 select, and the operations were conducted with the greatest care. 

 The absence of ferrous iron indicates a theoretically pure asgirine;: 

 returns of small quantities of ferrous iron betoken intermediate 

 stages between segirine and other pyroxenes, and substantiate 

 conclusions evinced by the microscopically zoned characters. 



The only other constituent deserving special mention is the 

 Zr0 2 which appears to be definitely combined with the silicate 

 as a Lavenite-like molecule. Inclusions of zircon in the segirine 

 were rare, and, taken together with the fact that the three 

 separate determinations proved its presence in fairly constant 

 amount, lend support to the above decision. 



Discussion of the formula. — In an attempt to develop a formula 

 in agreement with the analysis, zirconia was regarded as an acid 

 radicle from analogy with the cases of the closely related minerals 

 Lavenite and Wohlerite. Likewise, as in titaniferous augites, 

 the titania was taken as an acid radicle. 



The role of the water was not so evident, though it was 

 expected to rank with the alkalies, as is to be inferred in the 

 cases of several of the analyses quoted by Dana. 



The impossibility of meeting the demands of the analysis with 

 any concise formula is most certain evidence that, as was suspected 

 after the petrological examination, the pyroxene material selected 

 was not of uniform composition. 



The theoretical empirical formula NaFe(Si0 3 ) g given by Dana 

 is no doubt correct for the pure mineral, though only roughly 



